Back

On the assessment of more reliable COVID-19 infected number: the italian case.

Tradigo, G.; Guzzi, P. H.; Veltri, P.

2020-03-27 health informatics
10.1101/2020.03.25.20043562 medRxiv
Show abstract

COVID-19 (SARS-CoV-2) is the most recent pandemic disease the world is currently managing. It started in China at the end of 2019, and it is diffusing throughout Italy, one of the most affected countries, and it is currently spreading through European countries and USA. Patients affected by COVID-19 are identified employing medical swabs applied mainly to (i) citizens with COVID-19 symptoms such as flu or high temperature, or (ii) citizens that had contacts with COVID-19 patients. A percentage of COVID-19 affected patients needs hospitalisation, whereas a portion needs to be treated in Intensive Care Units (ICUs). Nevertheless, it is a matter of current intuition that COVID-19 infected citizens are more than those detected, and sometime the infection is detected too late. Thus there are many efforts in both tracking people activities as well as diffusing low cost reliable COVID-19 tests for early detection. Starting from mortality rates of diseases caused by viruses in the same family (e.g. MERS, SARS, H1N1), we study the relations between the number of COVID-19 infections and the number of deaths, through Italian regions. We thus assess several infections being higher than the ones currently measured. We thus focus on the characterisation of the pandemic diffusion by estimating the infected number of patients versus the number of death. We use such an estimated number of infections, to foresee the effects of restriction actions adopted by governments to constrain virus diffusion. We finally think that our model can support the healthcare system to react when COVID-19 is increasing.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.